Adalimumab and dexamethasone for treating non-infectious uveitis [ID763] Multiple Technology Appraisal 2 nd meeting: 12 th April 2017 Committee C Slides for Committee, projector and public [NoACIC] 1
The technologies Dexamethasone intravitreal implant Ozurdex (Allergan) 0.7 mg intravitreal implant (biodegradable) in an applicator Corticosteroid that inhibits proinflammatory mediators 870.00 per implant (BNF Dec 2016) 6 monthly cost: 870 (source: AR) Marketing authorisation Treatment of adult patients with inflammation of the posterior segment of the eye presenting as non-infectious uveitis. Adalimumab subcutaneous injection Humira (Abbvie) 40mg/0.4ml solution for injection every other week Monoclonal antibody that inhibits the pro-inflammatory cytokine, tumour necrosis factor (TNF)-alpha 704.28 per two prefilled pens/syringes or vials (BNF Dec 2016) 6 monthly cost: 4,578 (source: AR) Marketing authorisation Treatment of non-infectious intermediate, posterior and panuveitis in adult patients who have had an inadequate response to corticosteroids, in patients in need of corticosteroid sparing, or in whom corticosteroid treatment is inappropriate. 2
Treatment pathway Systemic pathway for patients with: Bilateral + active systemic Unilateral + active systemic Bilateral + no active systemic (via either pathway) Local pathway for patients with: Unilateral or asymmetric bilateral + no active systemic Bilateral + no active systemic (via either pathway) 1 st line: systemic steroids DEX licensed 1 st line: periocular steroids (may repeat) DEX licensed 2 nd line: Immunosuppressants (may also continue steroids 7.5mg/d): One: mycophenolate mofetil (or methotrexate) Two: mycophenolate mofetil (or methotrexate) + tacrolimus (or cyclosporine) VISUAL trials DEX and ADA licensed 3 rd line: Anti-TNF s (adalimumab, infliximab, etanercept) DEX and ADA licensed 2 nd line: Dexamethasone implant (may repeat) HURON trial DEX licensed Recreated using Figure 2 in Assessment Report. 3
Trial results from company submissions Adalimumab vs. placebo (95% confidence interval) Dexamethasone vs. sham (95% CI) Outcome VISUAL I VISUAL II HURON Time to treatment 0.50 0.57 Not reported failure*: HR (0.36 to 0.70) (0.39 to 0.84) Vitreous haze=0: RR (week 8* and 26) Mean change in visual acuity: MD Not reported Not reported Week 8*= 4.0 (2.0 to 7.6) Week 26= 2.2 (1.1 to 4.1) -0.07 (-0.11 to -0.02) -0.04 (-0.08 to 0.01) MD Not reported, p=0.002 at week 26 Vitreous haze (VH)**: MD -0.27-0.13 Week 8: -0.97, p<0.001 (-0.43 to -0.11) (-0.28 to 0.01) Week 26: -0.58, p<0.001 Macular oedema (change in Not reported Not reported Week 8: -87.0 (-147 to - thickness μm): MD 27) Week 26: -14.7 (-66 to 37) Visual Functioning Questionnaire-25 (VFQ- 25) (composite): MD 4.20 (1.02 to 7.38) 2.12 ( 0 84 to 5 08) MD Not reported, p=0.001 Abbreviations: HR hazard ratio; RR relative risk; MD mean difference *Primary outcomes; ** VISUAL: Mean change in VH; HURON: Mean VH score 4
Assessment group (AG) report Companies did not submit cost effectiveness models Indirect comparison not appropriate (for example differences in baseline therapy, unknown bilateral uveitis in HURON, differences in rescue therapy in HURON) AG Markov model 3 separate analyses based on evidence (active and inactive disease for adalimumab and active disease for dexamethasone) Can t distinguish bilateral and unilateral (>90% in VISUAL trials with bilateral disease) 5
AG Model Source: Figure 7 in AG report Patients receiving dexamethasone: assumed to have active disease Patients receiving adalimumab: active and inactive (assessed separately) Rates of transition defined from trials (apart from remission and blindness) Rate of blindness 0.0066 (Dick et al 2016) in base case Blindness utility 0.38 (Czoski-Murray 2009) in base case and 0.57 (Brown et al 1999) in exploratory analyses QoL in each health state from trials (EQ-5D for adalimumab, VFQ-25 for dexamethasone & AG mapped to EQ- 5D). Both assumed to capture QoL associated with AE during treatment Base case assumes no one in remission (exploratory analyses) 6
AG Base case Base case Total Inc. Inc. Total costs QALYs QALYs costs ICER Active uveitis dexamethasone Clinical practice 14.613 39,655 - - - Dexamethasone plus clinical practice 14.641 40,235 0.029 580 20,058 Active uveitis adalimumab Clinical practice 14.919 47,186 - - - Adalimumab plus clinical practice 15.110 65,401 0.191 18,215 95,506 Inactive uveitis adalimumab Clinical practice 15.244 48,111 - - - Adalimumab plus clinical practice 15.361 85,462 0.116 37,351 321,405 Deterministic results 7
AG exploratory analyses 1) Background blindness rate 0.0066 (Dick et al 2016) used in AG base case 0.0038 (Tomkins-Netzer et al 2014) used in exploratory analyses 0.0374 (Durrani et al 2004) UK based study used in exploratory analyses 2) Relative risk of blindness for dexamethasone Vary the relative risk of blindness from 0 (no one goes blind while on treatment) to 1 (blindness rate same as comparator) Base case 0.5 (blindness rate 50% lower in DEX vs. comparator) 3) Remission for adalimumab (not included in base case) Assume after 2 years on ADA (stable disease) some patients stop treatment because in remission and maintain same benefits AG exploratory analyses include annual rate either 0, 0.05, 0.1, 0.2 or 1 8
Committee preferred assumptions Model parameter Committee preferred assumption Rate of blindness 0.0066 acceptable for unilateral disease 0.0374 where higher risk of blindness (bilateral disease with macular oedema as proxy) Rate of remission Cost blindness Likely to be some remission after treatment with adalimumab < 7,700 per year Blindness utility 0.57 for all patients (Brown et al 1999) 9
AG exploratory analyses: adalimumab Rate of ICER (no blindness before treatment failure) remission* Blindness rate of 0.0374 (Durrani et al 2004) 0 33,003 0.05 25,171 0.1 20,821 0.2 15,994 1 6,942 Blindness rate of 0.0066 (Dick et al 2016) 0 95,506 0.05 77,414 0.1 67,363 0.2 56,214 1 35,299 *Annual rate of patients going into remission and discontinuing treatment whilst maintaining the benefit, if remaining on treatment at 2 years 10
AG exploratory analyses: dexamethasone Annual blindness rate ICER when varying relative risk of blindness from 0 (no blindness before treatment failure) to 1 (no effect) RR =0 RR=0.25 RR=0.5* RR=0.75 RR=1 Blindness utility of 0.38 (base case, Czoski-Murray et al. 2009 ) 0 (no blindness) 48,937 48,937 48,937 48,937 48,937 0.0066 a* 8,688 13,314 20,058* 30,805 50,627 0.0038 b 17,100 21,816 28,089 36,844 49,915 0.0374 c Dominates Dominates 557 10,900 56,329 Blindness utility of 0.57 (Brown et al. 1999) 0 (no blindness) 48,937 48,937 48,937 48,937 48,937 0.0066 a* 12,108 17,782 25,257* 35,550 50,627 0.0038 b 22,015 26,972 32,988 40,440 49,915 0.0374 c Dominates Dominates 853 15,198 56,329 High cost of blindness (upper bound of 95%) 0 (no blindness) 48,937 48,937 48,937 48,937 48,937 0.0066 a* 6,283 11,174 18,305* 29,668 50,627 0.0038 b 15,195 20,185 26,822 36,085 49,915 0.0374 c Dominates Dominates Dominates 8,534 56,329 11 *base case; Deterministic Dick al results 2016; b Tomkins-Netzer et al 2014; c Durrani et al 2004
ACD: Preliminary recommendations for adalimumab 1.1 Adalimumab is recommended as an option for treating non-infectious uveitis in the posterior segment of the eye in adults with inadequate response to corticosteroids, only if there is: active disease, that is, current inflammation in the eye macular oedema inadequate response to immunosuppressants systemic disease or both eyes are affected and worsening vision with a risk of blindness. 12
ACD: Preliminary recommendations for adalimumab (2) 1.2 Stop adalimumab for non-infectious uveitis in the posterior segment of the eye in adults with inadequate response to corticosteroids if there is 1 of the following: new active inflammatory chorioretinal or inflammatory retinal vascular lesions or both Failure to reduce anterior chamber cell grade of 0.5+ or less Failure to reduce vitreous haze grade of 0.5+ or less worsening of best corrected visual acuity by 3 or more lines or 15 letters. 13
ACD: Preliminary recommendations for dexamethasone 1.3 Dexamethasone intravitreal implant is recommended as an option for treating non-infectious uveitis in the posterior segment of the eye in adults, only if there is: active disease, that is, current inflammation in the eye and macular oedema. 14
ACD conclusions: clinical effectiveness Section ACD conclusion 4.2 In clinical practice, treatment depends on whether disease is: active (that is, current inflammation in the eye) or inactive (that is, limited inflammation, usually because of treatment with corticosteroids or immunosuppressants) systemic (when disease is not only in the eye) or non-systemic (when disease is limited to the eye) unilateral (when 1 eye is affected) or bilateral (when both eyes are affected). 4.6 Useful to distinguish unilateral from systemic and bilateral disease - people with higher risk of blindness are clinically important subgroup. 4.7 Both adalimumab and dexamethasone are clinically effective treatments for improving visual acuity, anterior chamber cell grade and vitreous haze. 15
ACD conclusions: cost effectiveness Section ACD conclusion 4.19 Adalimumab for inactive disease all ICERs > 80,000 per QALY gained Adalimumab unlikely to be used for inactive disease in clinical practice 4.20 Adalimumab for active disease More cost effective where high risk of permanent blindness (bilateral disease with macular oedema - useful proxy) ICERs around 33,000 per QALY gained (probably lower as rate of blindness underestimated for progressive loss of visual acuity). Reasonable to assume some patients in remission Stopping rule reflects strict criteria for treatment failure (VISUAL I) 4.21 Dexamethasone for active disease In practice dexamethasone generally used for unilateral disease ICERs ranged between 25,000 (treat better seeing eye-real risk blindness) and 49,000 per QALY gained (no risk of bilateral blindness but likely to be a significant overestimate as disutility of monocular blindness not modelled) Committee concluded that the ICER was likely to be within the range normally considered cost-effective 16
Consultation comments Companies for adalimumab (AbbVie) and dexamethasone (Allergan) Royal college of ophthalmologists (RCO), RNIB, Olivia s Vision (OV), Bird Shot Uveitis Society (BUS), Clinical expert (SS) NHS England (NHSE), Health improvement Scotland (HIS), Department of Health and AG 1 web response (NHS professional-clinical ophthalmology) 17
1. Recommendation changes (ADA) Comment from: RCO, OV, HIS, NHSE, AG Recommendations 1.1 to 1.3 need clarification treat based on all criteria or some? add intolerance to immunosuppressants? 18
2. Macular oedema as proxy for blindness? ADA and DEX Comment from: RCO, RNBP, Allergan, AbbVie, HIS, NHSE, AG, BUS, SS and web Committee concluded people with bilateral disease and macular oedema as a proxy for those at higher risk of permanent legal blindness Macular oedema should not be essential criterion for DEX and ADA for ADA captured by worsening vision with a risk of blindness for both ADA and DEX macular oedema is not the only proxy for blindness 19
2. Macular oedema as proxy for blindness? Dexamethasone Allergan provide ICERS based on longer re-treatment time (from post authorisation CONSTANCE long term safety study) of 44 weeks (30 weeks in base case) to support removing macular oedema as criterion Small risk blindness without macular oedema-likely to be cost effective Model changes ICER vs limited current practice Allergan (44 wks*) AG (30 wks*) 0.57 utility for blindness 14,016 25,257 0.57 utility and no risk of blindness 30,898 48,937 0.57 utility for blindness and low risk blindness (0.0038) 19,658 32,988 *Allergan use 44 weeks as DEX treatment effect vs. 30 weeks in AG scenario 20
2. Macular oedema as proxy for blindness? AG response Company increase duration of dexamethasone and change blindness to lower rate used in AG exploratory analyses (0.0038) and to 0 suggesting it is somewhere between these values AG able to reproduce company s ICERs (technically correct) AG question use of CONSTANCE study (no justification over other evidence) Papers used for risk of blindness likely to include patient with and without macular oedema. Effectiveness data in model does not differentiate between 2 groups (committee to decide if appropriate to consider patients without MO separately) 21
3. Adalimumab for unilateral disease? Comment from: RCO, OV, HIS, NHSE, BUS, SS Recommendation for adalimumab shouldn t restrict to bilateral disease occasions where used for unilateral disease (all other criteria fulfilled - local therapy contraindicated/failed or better seeing eye). If uveitis is in better seeing eye, risk of blindness could be as high as bilateral (and therefore, cost-effectiveness the same) 22
4. Stopping rule based on trial too prescriptive? (ADA) Comment from: RCO, RNBP, OV, AbbVie, HIS, NHSE, BUS, SS and web Treatment failure in trial not same as clinical practice Single flare of inflammation does not justify stopping adalimumab Suggested using stopping rule from NHS England interim policy 23
5. Use of DEX Comment from: RCO, RNIB, OV, Allergan, AG, BUS and web Note: issues not included in recommendations but comments suggest clarification in FAD is needed More than 3 implants may be used in practice SmPC: Very limited information on repeat dosing intervals less than 6 months. Currently no experience of repeat administrations Possible to treat bilateral disease (e.g. where no systemic disease or response to previous treatment, flare up in one eye) 24
Key Issues 1. Add intolerance to immunosuppressant? 2. Is macular oedema needed as a proxy? 3. Adalimumab for some unilateral disease (e.g. better seeing eye)? 4. Is stopping rule for adalimumab clinically appropriate? Potential equality issues in ACD consultation responses 25
Proposed changes to recommendations Current recommendation (proposed deletions in red and additions in green) 1.1 Adalimumab is recommended only if there is: active disease, that is, current inflammation in the eye and macular oedema inadequate response or intolerance to immunosuppressants and systemic disease or both eyes are affected or the better seeing eye is affected and worsening vision with a risk of blindness (for example from macular oedema). 1.2 Stop adalimumab if there is 1 of the following: new active inflammatory chorioretinal or inflammatory retinal vascular lesions or both failure to reduce anterior chamber cell grade of 0.5+ or less failure to reduce vitreous haze grade of 0.5+ or less worsening of best corrected visual acuity by 3 or more lines or 15 letters. 1.3 Dexamethasone is recommended only if there is: active disease, that is, current inflammation in the eye and worsening vision with a risk of blindness (for example from macular oedema) 26